PVL is the major pathologic substrate for neurological morbidity observed in premature infants. Hypoxia-ischemia and maternal-fetal infection are considered two major risk factors for development of PVL. In this project our overall hypothesis is that both hypoxia-ischemia and infection result in heightened innate immunity that in turn leads to injury of developing oligodendrocytes and neurons/axons. We propose that there are two major mechanisms of innate immune activation in PVL: (1) innate immune activation due to infection occurs through pathogen associated molecular patterns (PAMPs) and (2) innate immune activation due to hypoxia-ischemia occurs due to release of heat shock protein 60 (HSP60). In previously published reports and in preliminary data, we have shown that specific PAMPs activate the resident innate immune cells of the CNS (microglia) through toll-like receptors (TLRs). Activation of innate immune cells by PAMPs results in injury to developing oligodendrocytes and neurons/axons in CNS cultures. We have shown that HSP60 is a putative endogenous TLR4 agonist. HSP60 induces TNFalpha and nitric oxide production by microglia from wild-type but not TLR4 or MyD88 mutant mice. HSP60 is released from cells undergoing experimental necrosis or apoptosis. HSP60, like LPS, causes severe injury to axons/neurons in CNS cultures. Since axonal development may be inhibited in PVL, we have also examined mechanisms by which cell injury in PVL could lead to inhibition of axonal growth. We have found that neurons express TLR3 and that double stranded RNA, a TLRS ligand, inhibits neurite extension and causes growth cone collapse. This is the first example of a TLR functioning in neurons. Our preliminary data support the hypothesis that a convergence of innate immune activating mechanisms acting through toll-like receptors on microglia and neurons leads to the oligodendrocyte and axonal defects characteristic of PVL. We will test this hypothesis through the following specific aims: Aim 1. To characterize the developmental expression of TLRs in normal human white matter and in PVL. Hypothesis: TLR expression will be up-regulated in fetal versus white matter and in PVL lesions. Aim 2. To determine whether endogenous activators of microglia cause pre-OL injury in vitro. Hypothesis: (i) HSP60 activates microglia in a TLR4-MyD88 dependent fashion leading to secondary (bystander) injury of pre-OLs. Aim 3. To characterize the interaction of HSP60 with TLR4. Hypothesis: A restricted portion of the HSP-60 primary sequence interacts with TLR4; this interaction is required for activation of TLR4. Aim 4. To determine whether blocking microglial activation will protect against neonatal brain injury in models of PVL. Hypothesis: Specific inhibition of the TLR/MvD88 pathway will improve outcomes in mouse models of PVL.